Article Text

2018 EULAR recommendations for physical activity in people with inflammatory arthritis and osteoarthritis
  1. Anne-Kathrin Rausch Osthoff1,2,
  2. Karin Niedermann1,
  3. Jürgen Braun3,4,
  4. Jo Adams5,
  5. Nina Brodin6,7,
  6. Hanne Dagfinrud8,
  7. Tuncay Duruoz9,
  8. Bente Appel Esbensen10,11,
  9. Klaus-Peter Günther12,
  10. Emailie Hurkmans13,
  11. Carsten Bogh Juhl14,15,
  12. Norelee Kennedy16,
  13. Uta Kiltz3,4,
  14. Keegan Knittle17,
  15. Michael Nurmohamed18,
  16. Sandra Pais19,
  17. Guy Severijns20,
  18. Thijs Willem Swinnen21,22,
  19. Irene A Pitsillidou23,
  20. Louise Warburton24,
  21. Zhivko Yankov25,
  22. Theodora P M Vliet Vlieland2
  1. 1 School of Health Professions, Institute of Physiotherapy, Zurich University of Applied Sciences, Winterthur, Switzerland
  2. 2 Department of Orthopaedics, Rehabilitation and Physical Therapy, Leiden University Medical Center, Leiden, Netherlands
  3. 3 Rheumazentrum Ruhrgebiet, Herne, Germany
  4. 4 Ruhr University, Bochum, Germany
  5. 5 Faculty of Health Sciences and Arthritis Research UK Centre of Excellence for Sport, Exercise and Osteoarthritis, University of Southampton, Southampton, UK
  6. 6 Department of Orthopaedics, Danderyd University Hospital Corp., Stockholm, Sweden
  7. 7 Department of Neurobiology, Care Sciences and Society, Division of Physiotherapy, Karolinska Institutet, Huddinge, Sweden
  8. 8 National Advisory Unit on Rehabilitation in Rheumatology, Diakonhjemmet Hospital, Oslo, Norway
  9. 9 PMR Department, Rheumatology Division, Marmara University, School of Medicine, Istanbul, Turkey
  10. 10 Copenhagen Center for Arthritis Research, Center for Rheumatology and Spine Diseases, Centre for Head and Orthopaedics, Rigshospitalet., Glostrup, Denmark
  11. 11 Department of Clinical Medicine, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark
  12. 12 University Center of Orthopaedics and Traumatology, University Medicine at Technische Universität Dresden, Dresden, Germany
  13. 13 Department Social Affaire and Health, ECORYS Nederland BV, Rotterdam, Netherlands
  14. 14 Department of Sports Science and Clinical Biomechanics, University of Southern Denmark, Odense, Denmark
  15. 15 Departmentof Occupational and Physical Therapy, University of Copenhagen, Herlev and Gentofte Hospital, Copenhagen, Denmark
  16. 16 School of Allied Health, Faculty of Education and Health Sciences and Health Research Institute, University of Limerick, Limerick, Ireland
  17. 17 Department of Social Psychology, Faculty of Social Sciences, University of Helsinki, Helsinki, Finland
  18. 18 Department of Rheumatology, Amsterdam Rheumatology and immunology Center, VU University Medical Center, Amsterdam, The Netherlands
  19. 19 Centre for Biomedical Research, University of Algarve, Faro, Portugal
  20. 20 EULAR PARE Patient Research Partner, ReumaNet, Leuven, Belgium
  21. 21 Department of Development and Regeneration, Skeletal Biology and Engineering Research Center, KU Leuven, Leuven, Belgium
  22. 22 Division of Rheumatology, University Hospitals Leuven, Leuven, Belgium
  23. 23 EULAR Patient Research Partner, Cyprus League Against Rheumatism, Nicosia, Cyprus
  24. 24 Primary Health Care Sciences, Keele University, Keele, UK
  25. 25 PRP (Patient Research Partner EULAR), Bulgarian Ankylosing Spondylitis Patient Society, Sofia, Bulgaria and ASIF (Ankylosing Spondylitis International Federation), London, UK
  1. Correspondence to Professor Karin Niedermann, School of Health Professions, Institute of Physiotherapy, Zurich University of Applied Sciences, Winterthur, Switzerland; karin.niedermann{at}


Regular physical activity (PA) is increasingly promoted for people with rheumatic and musculoskeletal diseases as well as the general population. We evaluated if the public health recommendations for PA are applicable for people with inflammatory arthritis (iA; Rheumatoid Arthritis and Spondyloarthritis) and osteoarthritis (hip/knee OA) in order to develop evidence-based recommendations for advice and guidance on PA in clinical practice. The EULAR standardised operating procedures for the development of recommendations were followed. A task force (TF) (including rheumatologists, other medical specialists and physicians, health professionals, patient-representatives, methodologists) from 16 countries met twice. In the first TF meeting, 13 research questions to support a systematic literature review (SLR) were identified and defined. In the second meeting, the SLR evidence was presented and discussed before the recommendations, research agenda and education agenda were formulated. The TF developed and agreed on four overarching principles and 10 recommendations for PA in people with iA and OA. The mean level of agreement between the TF members ranged between 9.8 and 8.8. Given the evidence for its effectiveness, feasibility and safety, PA is advocated as integral part of standard care throughout the course of these diseases. Finally, the TF agreed on related research and education agendas. Evidence and expert opinion inform these recommendations to provide guidance in the development, conduct and evaluation of PA-interventions and promotion in people with iA and OA. It is advised that these recommendations should be implemented considering individual needs and national health systems.

  • rehabilitation
  • rheumatoid arthritis
  • spondyloarthritis
  • Hip osteoarthritis
  • knee osteoarthritis

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Physical activity (PA) is defined as ‘any bodily movement produced by skeletal muscles that results in energy expenditure above resting (basal) levels. PA broadly encompasses exercise, sports and physical activities done as part of daily living, occupation, leisure and active transportation’.1 2 Exercise is a subcategory of PA ‘that is planned, structured and repetitive and [that] has, as a final or intermediate objective, the improvement or maintenance of one or more dimensions of physical fitness’.1 2 PA-interventions can be provided or performed individually or in groups, supervised or non-supervised, in acute or chronic health states, but should always include behavioural change techniques (BCT) to promote long-term adherence.3 4

To promote the health benefits of PA in the general population, the WHO5 and American College of Sports Medicine (ACSM)2 have provided internationally accepted recommendations for PA (table 1). In this manuscript, the term PA always includes both physical activity and exercise according to the definitions above.

Table 1

Public Health recommendations for PA

Inflammatory arthritis (iA, in this manuscript encompassing rheumatoid arthritis (RA) and spondyloarthritis (SpA)) and osteoarthritis (OA) (in this manuscript encompassing hip/knee OA (HOA/KOA)) are major causes of pain and disability worldwide.6 There is strong evidence for the benefits of PA on improvements on disease activity,7 activities and participation; however, people with rheumatic and musculoskeletal diseases (RMDs) are in general less active compared with healthy controls.8–10 Possible underlying reasons could be that healthcare providers (HCP, including rheumatology health professionals (eg, physiotherapist (PT), occupational therapist (OT), nurse, podiatrist, psychologist), physical education professions and medical doctors (rheumatologists and other specialists)) and people with iA and OA may be reluctant towards engaging in PA, fearing flare-up or joint damage by exercising.11 Furthermore, current clinical management recommendations such as the European League Against Rheumatism (EULAR) recommendations on the management of RA,12 SpA13 or HOA/KOA14 and the ACSM guidelines for exercise testing and prescription15 recommend exercise and/or PA, but none of these is specific regarding the required type and dosage. Therefore, it is not clear how these recommendations should be used in routine clinical care. In particular, the evidence on the effectiveness and safety of exercise and PA to a level that meets public health (PH) recommendations has not yet been clearly examined and defined in people with RMDs. A EULAR task force (TF) was therefore set up (1) to evaluate if the PH recommendations for PA are applicable for people with iA and OA; (2) to develop evidence-based recommendations on PA-promotion and -delivery in the management of people with iA and OA and (3) formulate an educational and research agenda.

These EULAR recommendations for PA in people with iA and OA are for HCPs, patient organisations and policy makers.


The EULAR standardised operating procedures for the development of recommendations were followed.16 The AGREE II-instrument17 was used to structure this manuscript.

The multidisciplinary TF consisted of a selection of 22 European PA-experts (six medical doctors, including three rheumatologists, one of them specialised in cardiovascular diseases, one GP), one orthopaedic surgeon; nine PTs, a psychologist, an OT, a nurse and a human movement scientist) and three patient representatives. A steering group managed the process (convenor KN, methodologist TVV, expert JB, fellow AR).

During the first TF meeting, definitions of exercise and PA were clarified and the TF agreed to follow the ACSM position stand.2 The TF agreed that RA and SpA as predominant iA conditions, and HOA/KOA as most relevant for PA recommendations would represent the field of iA and OA, respectively. Clinically relevant questions on the provision of advice and guidance regarding exercise and PA, from which 13 research questions were defined by consensus to guide the subsequent detailed systematic literature review (SLR) (online supplementary table S1).

Supplementary file 1

Two SLRs were performed by AR with the support of two librarians and under the supervision of the convenor and methodologists. The questions were written according to the Population, Intervention, Comparison, Outcome (PICO) format,18 resulting in two PICOs: (1) on effectiveness, safety and feasibility of PA and (2) on facilitators and barriers towards PA (online-supplementary table S2). For the first PICO, the fellow searched for key meta-analyses (MAs) or systematic reviews (SRs) including randomised controlled trials (RCTs) that investigated the effectiveness of PA-interventions in adults with RA/SpA/HOA/KOA. The SLR was performed in PubMed/Medline, Cochrane Library, Embase, Web of Science, Emcare and PsycInfo, using both MeSH terms and freetext, covering the time frame until 4/2017. For the second PICO, a SLR, covering the time frame until 7/2017, was performed in PubMed/Medline and Cochrane Library including qualitative studies if they described facilitators and barriers regarding PA (including exercise) in people with RA/SpA/HOA/KOA. Experts in the field of RA (EH), SpA (HD), OA (CJ) and behaviour change (KK), respectively, checked if all relevant titles and abstracts were included.

Supplementary file 2

Based on the PICOs, the same author (AR) screened the titles and abstracts according to inclusion and exclusion criteria. Potentially relevant articles were identified and full text versions evaluated. Studies including adults (>18 years) with RA/SpA/HOA/KOA that included PA interventions that met the PH recommendations according to the ACSM principles2 regarding frequency, intensity and duration for effective interventions were eligible for inclusion. All data extractions were checked by experts from the TF.

Studies measuring the effectiveness of PA-interventions were meta-analysed. These results and detailed descriptions of the methods are reported elsewhere.19 Studies were used for answering more than one research question if appropriate. For clinical studies evaluating the effectiveness of PA, the Cochrane Risk of Bias Assessment Tool was used to assess selection bias, performance bias, detection bias, attrition bias and reporting bias20 by two independent assessors (AR, CH). An additional person (KN) helped to resolve any differences in rating between the assessors. The research evidence was categorised according to the Oxford levels of evidence.21

During the second TF meeting, the results from the SLR were presented, and the experts developed ‘overarching principles’ (background statements to preface recommendations) and drafted 10 recommendations through an iterative process of discussion and consensus. After the meeting, the recommendations were collated and sent to the TF members by email, to rate the level of agreement (LoA) independently and anonymously on a 0–10 point scale (0=totally disagree, 10=totally agree). Mean LoA >8 would be considered a ‘high’ LoA. Furthermore, the TF formulated a research agenda and education agenda based on identified gaps in the evidence.


The search yielded 3471 references, 96 of which were included in the SLR: Four MA/SR7 22–24 and 66 RCTs25–93 investigated the effects of exercise interventions, 11 RCTs94–106 investigated the effects of a PA-promotion-intervention, 11 qualitative studies and literature reviews3 11 107–115 described barriers and facilitators regarding PA (figure 1A,B). The included RCTs were published between 1985 and 2017. Most information is from studies with low (48%) or unclear (39%) risk of bias (online-supplementary figure S1).

Supplementary file 3

Figure 1

Flowcharts of the literature search related to PICO_1 (A) and PICO_ 2 (B). ACSM, American College of Sports Medicine; OA, osteoarthritis; PA, physical activity; PICO, Population, Intervention, Comparison, Outcome; RA, rheumatoid arthritis; RCT, randomised controlled trial; SpA, spondyloarthritis; SR, systematic review.

The TF agreed on four overarching principles and 10 recommendations for PA in people with RA/SpA/HOA/KOA based on SLR and expert opinion. High loA was achieved for 9 out of 10 recommendations and 2 recommendations were graded as strength level A. Table 2 summarises the overarching principles and recommendations with their associated level of evidence, strength of recommendation and LoA.

Table 2

Recommendations for PA and exercise in people with inflammatory arthritis and OA

Recommendation 1: PA as integral part of standard care

Given the evidence for effectiveness, feasibility and safety, the PH recommendations for PA are applicable, and thus, PA should be an integral part of standard care for people with RA/SpA/HOA/KOA. PA according to PH recommendations2 is effective on PA level, physical fitness as well as disease-specific and general outcomes in people with RA/SpA/HOA/KOA (category l evidence16). Our MA including 16 RCTs26 35 36 42 43 50 54 56 57 61 70 showed that cardiovascular exercises have a moderate beneficial effect on cardiovascular fitness (evaluated in VO2 max) in all three conditions. Our MA including 25 RCTs25 28 31 34 38 39 44 47 49–51 59 62–66 72 75–78 81–83 85 86 88 90 91 showed that muscle strength exercises have a moderate beneficial effect for muscle strength in people with RA and HOA/KOA. Our MA including seven RCTs52 55 58 78 88 90 116 showed that combined exercises (aerobic or strength exercises plus flexibility exercises) had no effect on flexibility in people with SpA or HOA/KOA. However, exercise conditions, assessments and outcome measures varied greatly. There is no study comparing the effect of flexibility exercises alone versus no exercises. In one RCT,48 the effect of a neuromotor-exercise programme on neuromotor performance was investigated in people with RA showing a positive effect. Eleven RCTs described the promotion of daily PA. Our MA including six RCTs95 98 101 102 104 117 applying BCTs for the counselling intervention showed a small beneficial effect.

Feasibility of interventions can be captured by adherence to the intervention or the study protocol.118 Adherence to interventions (number of sessions attended/total number of sessions) has been reported in 26 RCTs (35%) and the mean adherence was 69% in people with SpA, 71% in people with RA and 79% in people with HOA/KOA. However, the (self-) reported adherence to intervention might be overestimated due to recall bias or social desirability. In 68 RCTs (94%), protocol violations were reported, with approximately 10% of these being disease-related or intervention-related.

PH recommendations for PA can be considered safe. No detrimental effects were reported, rather beneficial effects on disease activity and symptoms in iA.7 Forty-four per cent of all included RCTs reported on adverse events (AE), of those 62% described no AE and 38% describe minor AE such as transitional exercise related joint or muscle pain.

Recommendation 2: Responsibility for PA promotion

All HCPs should have a responsibility for PA promotion and collaborative working that facilitate a close cooperation between different professions to support appropriate disease management. This statement was based on the finding that 66% of the included studies reported the profession of the HCP providing the intervention, of which 75% were PTs.25 31 36 40 44 45 48 50 53 55 58 61 64–66 70 73–79 81 84 87 88 91 94 96 101–105 119 120 However, the functions and responsibilities of HCPs vary across Europe.121 122 Therefore, the TF agreed that PA advice should be provided by all HCPs.

Recommendation 3: Delivery of PA

The delivery of interventions should be performed by HCPs competent in the field of PA principles and rheumatic conditions. The reporting of training on PA guidelines was rare. One study59 described a ‘4 hours education session on cardiovascular training’, others described the instructing person as ‘trained’25 50 69 70 84 88 123 or ‘experienced’.31 49 76 77 88 Some studies with focus on the promotion of daily PA described training sessions on behaviour change skills like Motivational Interviewing.94 96 104

Recommendation 4: Evaluation of PA

The PA level (active or non-active) and the exercise domains (cardiorespiratory, muscle strength, flexibility and neuromotor) should be routinely assessed. Of 11 trials investigating the effect of PA promotion interventions, three RCTs94 96 105 described baseline screening to distinguish between active and non-active persons before starting the tailored PA-intervention. Specific tools are needed to assess each domain.15,p. 68

Recommendation 5: General and disease-specific contraindications

Tools for specific contraindications (CIs) were found;15 94 124 however, available general or national guidelines defining absolute or relative CIs should be followed as a priority.

Recommendation 6: Personalised aims and evaluation

The PA-interventions should be based on individual aims, which should be regularly evaluated. This can be done by PA assessments and any other assessments related to the individual aims. As PA assessments, performance–based tests, patient-reported outcome measures (eg, SQUASH,104 PASE94) and self-monitoring tools (eg, wearables such as Fitbit,100 pedometer99 or accelerometer101) were identified. However, we did not evaluate the validity and reliability of the assessments applied.

Recommendation 7: General and disease-specific barriers and facilitators

General and disease-specific barriers (that are not CI per se) and facilitators should be addressed as described in 11 studies.11 107–115 125 Disease-specific barriers included lack of knowledge about the disease, lack of knowledge about safe exercising (both in people with iA/OA and HCPs) and symptoms like pain, fatigue, stiffness, reduced mobility, fear of flare-ups or causing damage. Disease-specific facilitators included positive impact of exercise in symptoms or disease control, information about disease and correct exercising, the use medication for pain prior to exercising, using self-regulation techniques, supportive, but not controlling encouragement from HCPs and a supportive social background.

Recommendation 8: Individual adaptations to PA following individualised assessment

Adaptations to PA should be made on a comprehensive individual assessment. However, no evidence on the necessity of general adaptations in people with RA/SpA/HOA/KOA was found. In some RA studies the ‘24 hour-rule’ was applied, that is, the exercise intensity was reduced when the increased pain persisted for more than 24 hours.23 40 50 ACSM provides adaptations to exercise testing in people with arthritis (eg, no high-intensity testing if acute inflammation) and training such as exercising when pain is typically least severe or to train carefully in order to reduce risk of associated injuries, although no clear evidence that high-impact activities cannot be engaged during active inflammation.15, pp. 298–301 Individual disease-related barriers (eg, symptoms) may determine these adaptations.

Recommendation 9: Behaviour change techniques

BCTs should be an integral component of PA-interventions. Several behaviour change theories were used in PA promotion interventions in the field of RA and HOA/KOA,4 126 but the reporting was poor. Future research based on theories in design, evaluation and interpretation of findings is needed.

A meta-analysis of six RCTs94 95 98 101 102 104 investigating the effects of a PA promotion intervention according to general PA recommendations2 and based on counselling interventions that apply BCTs showed a small beneficial effect on PA level.19 Counselling interventions show a small beneficial effect if BCTs are applied.19

Recommendation 10: Modes of delivery

HCPs should consider the whole range of modes to deliver interventions. No evidence on the superiority of specific delivery modes was found. The delivery modes of PA-interventions vary considerably and are mostly described as ‘land-based and/or water-based’ and ‘supervised and individualised’, the latter usually applied to group settings. As booster strategies phone calls,36 96 98 105 devices (eg, pedometer,98 99 wearable100 101), home visits,63 70 log book,36 51 76 98 web-based instructions,127 written material,51 54 103 visual instructions (eg, video103) were reported.

Research and education agendas

Based on the gaps identified in the literature, the TF discussed and proposed a research agenda (box 1) with the prioritised research topics and an education agenda (box 2) with topics for education and training in PA promotion for HCPs. Evidence on impact of (reducing) sedentary behaviour emerged as an important future research topic.

Box 1

Research agenda for physical activity (PA) in people with inflammatory arthritis and osteoarthritis

  1. To evaluate the long-term effectiveness of PA at different intensities and types and monitoring of adverse events (AE).

  2. To evaluate links between PA behaviour and disease-specific outcomes.

  3. To evaluate the long-term effectiveness of sedentary behaviour reduction, including the monitoring of AE.

  4. To evaluate links between sedentary behaviour and disease-specific outcomes.

  5. To identify which PA-intervention strategies work best to increase PA level and adherence in various subgroups.

  6. To identify markers of response and non-response to PA treatment.

  7. To identify disease-specific contraindications on different exercise domains (cardiovascular, strength, flexibility, neuromotor).

  8. To further develop and evaluate strategies to reduce and monitor a change in sedentary behaviour.

  9. To develop PA-interventions targeting all exercise dimensions simultaneously with special focus on feasibility.

  10. To evaluate and recommend valid PA assessments feasible for the use in clinical practice.

  11. To study how to facilitate PA behaviour change immediately from screening onwards and how to address facilitators and barriers.

  12. To identify facilitators and barriers of healthcare providers towards applying the PA recommendations.

  13. To perform long-term effectiveness trials on combined interventions including other health behaviours.

Box 2

Education agenda for physical activity (PA) in people with inflammatory arthritis and osteoarthritis

  1. Increase knowledge about PA among health professionals (HPs), physicians and people with inflammatory arthritis and osteoarthritis.

  2. Increase HPs’ and physicians’ skills in communicating the role of PA in managing general health and disease-specific issues.

  3. Include knowledge and skills on PA promotion in all HPs’ and physicians’ undergraduate training curricula.

  4. Develop a EULAR training module on PA for HPs and rheumatologists.

  5. Propose a session on PA at every EULAR congress.

  6. Develop education materials for people with inflammatory arthritis and osteoarthritis.


The TF agreed on 4 overarching principles and 10 recommendations for PA in people with RA/SpA/HOA/KOA, which integrated the perspectives of the TF members from different professional, cultural and personal backgrounds. This led to a broad consensus on the principles and recommendations within the group and ought to foster its feasibility and practicability in the diverging health systems across Europe.

The LoA on the recommendations among the TF members was very high. The only exception was about the competency of HCP, which may be due to country specific differences in the availability of HCP competent in PA promotion.

Although the PH recommendations for PA are well established, the feasibility and applicability of these for people with iA and OA has not been assessed so far. Accordingly, the development of the recommendations was needed. Expectedly, they emphasise the importance of PA and will guide future PA-interventions in people with chronic rheumatic conditions.

PA promotion is a behavioural intervention and therefore BCT are central components in PA-interventions. Identifying effective and cost-effective BCT within PA promotion intervention in people with chronic conditions is currently a hot topic in research and for example a research priority of the National Institute for Health and Care Excellence, UK.128

We decided a priori to include only studies fulfilling the PH recommendations for PA according to ACSM principles.2 This was a far-reaching decision, which allowed drawing stronger conclusions on the effectiveness and especially the safety of correctly dosed PA-interventions. We followed a pragmatic search strategy with the plan to answer all RQs related to PICO 1 with findings of available SR/MA. However, there were no SR/MA on all exercise dimensions and all conditions available; this led to extracting single RCTs from high-quality SR/MA. This, however, excluded high-quality reviews (eg, Cochrane reviews) and RCTs that did not fulfil the ACSM principles and affected the potential to report 1A evidence according to Oxford levels of evidence.21 Furthermore, only one reviewer screened the abstracts and decided on unclear abstracts together with a second reviewer, which is not fully in line with standard procedures of a SLR.129 However, we applied a double-check by experts to ensure that no relevant studies were missed.

A major problem for data extraction and interpretation was that the reporting of interventions in most studies was incomplete. Manuscripts that applied TIDieR130 (Template for Intervention Description and Replication) guidelines reported more precisely the PA-interventions and substantially improved the objective evaluation of the PA-interventions.

For the research questions related to the effectiveness and safety of PA-interventions and BCT, the PICO scheme was applied, resulting in 1A level of evidence. All other research questions we had to answer in a descriptive way limiting the level of evidence to 3 to 4. However, this limitation is due to the nature of the research questions. Nevertheless, the qualitative studies may provide valuable insight into important PA-related fields, such as assessments, barriers and facilitators, PA promotion strategies.

The recommendations focused on the conditions RA/SpA/HOA/KOA, the most prevalent RMD conditions to increase the generalisability and applicability of the recommendations. However, large heterogeneity between these conditions may limit the precision of the recommendations. Therefore, additional disease-specific recommendations are desirable. In addition, not all subconditions were considered and represented (eg, juvenile arthritis).

The research agenda highlights several areas where scientific evidence is lacking. It is a clear ambition to implement these recommendations into daily clinical routine. Due to the different health systems across Europe, development and evaluation of target group and culture-specific implementation strategies are needed and should involve all stakeholders.


We thank the librarians Mrs José Plevier and Mr Jan W Schoones, Walaeus Library, Leiden University Medical Center, the Netherlands, for supporting our literature search, and Christian Horvath, Zurich University of Applied Sciences, Switzerland, MSc student for his help in the Cochrane risk of bias assessment.


Supplementary materials

  • Lay summary

    Disclaimer : This is a summary of a scientific article written by a medical professional (“the Original Article”). The Summary is written to assist non medically trained readers to understand general points of the Original Article. It is supplied “as is” without any warranty. You should note that the Original Article (and Summary) may not be fully relevant nor accurate as medical science is constantly changing and errors can occur. It is therefore very important that readers not rely on the content in the Summary and consult their medical professionals for all aspects of their health care and only rely on the Summary if directed to do so by their medical professional. Please view our full Website Terms and Conditions.
    Copyright © 2019 BMJ Publishing Group Ltd & European League Against Rheumatism. Medical professionals may print copies for their and their patients and students non commercial use. Other individuals may print a single copy for their personal, non commercial use. For other uses please contact our Rights and Licensing Team.


  • A-KRO and KN contributed equally.

  • Handling editor Josef S Smolen

  • Contributors AR and KN contributed equally. AR was the research fellow for the project, undertaking the SLR. The fellow was supervised by the steering group consisting of KN (convenor), TPMVV (methodologist), JB (expert). KN and TPMVV supervised the process of the SLR. KN organised and chaired the TF meetings. AR and KN drafted the manuscript with advice from TPMVV and JB. All authors have contributed to the recommendations by participating in the TF meetings; during discussion and agreement on the recommendations; revising and approving the manuscript for publication.

  • Funding The TF would like to thank EULAR for financial support of this work.

  • Competing interests None declared.

  • Patient consent Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.